Drug toxicity is a leading cause of acute liver failure. In the evaluation of hepatic failure, the clinical laboratory plays a vital role in diagnosis so that appropriate treatment can be initiated in a timely manner.
Acetaminophen (N-acetyl-p-aminophenol) has long been prescribed as an analgesic and antipyretic. It is widely available without prescription and is an active component in many common therapeutic formulations, such as cold and flu remedies. The widespread use of this drug places it high on the list of suspected hepatotoxic agents in patients presenting with liver malfunction.
While therapeutic doses of acetaminophen rarely cause adverse effects, cases have been reported where chronic, excessive use of acetaminophen has led to hepatotoxicity and nephrotoxicity. Ingestion of acute overdose quantities of acetaminophen causes a depletion of glutathione stores and accumulation of toxic metabolites in the liver, which can cause severe or even fatal liver failure.
When acetaminophen is ingested in excessive quantities, a highly reactive intermediate, N-acetyl-p-benzoquinoneimine, accumulates in the liver. This intermediate reacts with thiols in the liver, particularly glutathione. Glutathione is oxidized to glutathione disulfide (GSSG). Excessive levels of GSSG in the liver cause necrosis. Acetaminophen toxicity is generally reported at serum concentrations above about 20 mg/dL (1324 μmol/L).
The glutathione precursor, N-acetylcysteine (NAC), is often administered as an antidote for acetaminophen overdose. About 70% of NAC administered is metabolized in the liver. It is believed that NAC functions as an antidote for at least the following reasons: it is a precursor for glutathione, it is a powerful anti-oxidant, and it increases the efficiency of GSSG reductase in the liver. The administration of NAC is believed to minimize or prevent the damage caused by an overdose of acetaminophen, at least in part, by replenishing glutathione stores and preventing an accumulation of GSSG in the liver.
A high concentration of NAC is often administered in an initial loading dose followed by maintenance levels of NAC throughout the course of treatment. The loading dose can result in serum levels of NAC of 2000 mg/L or higher, and maintenance levels are often about 800 mg/L to 1000 mg/L. It is desirable to monitor acetaminophen levels throughout the course of NAC treatment to ensure an appropriate therapeutic level is maintained while avoiding unnecessary or excessive exposure to NAC.
The incidence of accidental, as well as intentional, acetaminophen overdose has increased significantly. The diagnosis and treatment of acetaminophen overdose requires early detection and accurate measurement of the drug in the body. The amount of acetaminophen on board must be quickly and accurately determined so that clinicians can rapidly administer an appropriate therapeutic dose of NAC to the patient. There is a high demand for rapid, reliable and robust clinical assays for determining acetaminophen concentration in biological samples.
Known methods for determining acetaminophen levels in biological samples include, for example, various chromatographic and spectrophotometric techniques.
Gas-liquid chromatography and high-performance liquid chromatography have proven to be reliable and accurate methods for determining acetaminophen levels in biological samples, however both are lengthy procedures that require expensive instrumentation and a high level of technical skill to perform. Such methods are not particularly suited for Stat laboratories, where rapid results are required.
Differential spectrophotometry has been widely used but this method requires time-consuming solvent extractions, which are undesirable in clinical assays. More rapid spectrophotometric methods generally fail to offer the desired specificity.
Colorimetric techniques include simple colorimetty as well as enzyme-based colorimetric assays. Various immune-based assays are also available but these tend to be significantly more expensive and therefore less desirable, particularly in a clinical setting.
While enzyme-based assays are convenient and economical compared to immune-based assays, they are generally less reliable in that they are prone to interference with biological molecules often present in patient samples, such as bilirubin and hemoglobin. Elevated levels of such molecules in patient samples can cause false positive results (see, for example, Bertholf et al., 2003), which can potentially lead to misdiagnosis and inappropriate choice or dose of treatment.
Known enzymatic assays are also subject to interference in the presence of therapeutic levels of NAC. Therefore, enzymatic assays cannot generally be used to monitor acetaminophen levels during the course of NAC treatment due to inaccuracy in the acetaminophen levels measured. This is a significant disadvantage of known enzymatic acetaminophen assays.
Known enzymatic assays employ three main components: an aryl acylamidase enzyme, a chromogenic (or color-forming) compound, and an oxidizing agent of sufficient oxidative potential to catalyze the coupling reaction.
Aryl acylamidase cleaves the amide bond of acetaminophen to yield p-aminophenol and acetate. The p-aminophenol is then reacted with the chromogenic compound in an oxidative coupling reaction in the presence of an oxidizing catalyst to form a colored product. Typical catalysts include metal salts or metal complexes of species having reactive oxygen or functional groups, such as permanganate, periodate, persulfate, sulfate, or acetate. The change in absorbance, typically measured at a wavelength that captures the peak absorbance of the colored product, is then used to determine the concentration of acetaminophen in the sample. This may be determined by comparing the absorbance values obtained against a standard or set of standards having known acetaminophen concentration and assayed by the same method. The number of moles of colored product formed is typically proportional to the number of moles of acetaminophen initially present in the sample.
The earliest enzyme-based acetaminophen assays required very long incubation times, often greater than 1 hour for each of the hydrolysis and oxidative coupling reactions, thus rendering them unsuitable for use in an emergency clinical setting.
Hammond et al. (1984) developed a rapid enzyme-based assay for determining acetaminophen concentration in serum using an aryl acylamidase to hydrolyze acetaminophen to p-aminophenol. The p-aminophenol is subsequently reacted with o-cresol in an oxidative coupling reaction catalyzed by copper sulfate, to form an indophenol dye. The change in absorbance at the peak wavelength of the dye (615 nm) is then used to determine acetaminophen levels. While this method provides rapid detection of acetaminophen, it is subject to significant interference in the presence of therapeutic levels of NAC and cannot be used reliably during NAC treatment. A similar method utilizing o-cresol in the presence of an oxidizing catalyst is prone to bilirubin interference (Bertholf et al., 2003), leading to false positive results in hyperbilinibinemic patients.
Morris et al. (1990) disclose an automated enzyme-based assay for measuring acetaminophen in a sample. Automated assays are generally preferred for clinical laboratories. The method uses an aryl acylamidase for hydrolysis of acetaminophen to p-aminophenol, followed by oxidative coupling with 8-hydroxyquinoline in the presence of manganese ions to form a blue product. The reagents are lyophilized for storage stability and must be reconstituted prior to use. Assays involving a reconstitution step are less desirable than liquid-stable assays and are more prone to error.
Known acetaminophen assays using 8-hydroxyquinoline or a derivative thereof as a chromophore are subject to interference in the presence of therapeutic levels of NAC (i.e. >800 mg/L). The present inventors tested two commercially available acetaminophen assays (Genzyme Diagnostics P.E.I. Inc., PEI, Canada) containing either 8-hydroxyquinoline-5-sulfonic acid (8-HQ5SA) or 8-hydroxyquino line hemisulfate (8-HQHS) as the chromophore. Although accurate acetaminophen measurements in the absence of NAC were seen, there was a significant (i.e. >about 10%) decrease in acetaminophen recovery in the presence of therapeutic levels of NAC. It was discovered that the presence of NAC affected the oxidative coupling reaction in the assay rather than the enzymatic conversion of acetaminophen to p-aminophenol. There was a considerable difference in recovery between the 8-HQ5SA and 8-HQHS assays, with the 8-HQ5SA assay being significantly more susceptible to NAC interference, indicating that even a slight difference in the chemical structure of the chromophore can be crucial to the coupling reaction when NAC is present.
Chen et at (2004) describe an assay for quantifying p-aminophenol in urine to assess exposure to aniline in the workplace. Urine p-aminophenol levels serve as a biological marker of aniline toxicity since about 15 to 60% of absorbed aniline is oxidized to p-aminophenol in vivo. The urine must be acidified and pretreated to release free p-aminophenol from the conjugated forms excreted in urine. The assay involves an oxidative coupling reaction using 2,5-dimethylphenol (p-xylenol) as the chromophore to form a colored product. The coupling reaction is catalyzed by sodium periodate, a strong oxidizer, to form a colored product. It was speculated that quantifying p-aminophenol levels in urine may be useful for assessing acetaminophen overdose, although this was neither explored nor demonstrated.
Afshari and Lui (2001) describe a non-enzymatic method for quantification of acetaminophen in serum. Free unconjugated acetaminophen is first separated from endogenous interferents by an extraction step followed by hydrolysis to p-aminophenol using heat (i.e. boiling for 10 minutes) and acid. This is a non-selective hydrolysis reaction compared to an enzymatic reaction. The hydrolysis reaction is followed by oxidative coupling of p-aminophenol to 2,5-dimethylphenol (p-xylenol) in the presence of sodium periodate, a strong oxidizer, to form a colored product. The need to extract the acetaminophen from the sample and boil the samples renders this method undesirable for use in an emergency clinical setting and also unsuitable for automation.
While enzymatic acetaminophen assays are convenient and more affordable than immuno-based assays, many clinical laboratories favor the immuno-based assays since they are unaffected by the presence of NAC in a sample. It is desirable to reliably measure acetaminophen levels during the course of NAC treatment. Immuno-based assays are also less susceptible to interference in the presence of biological molecules, such as bilirubin and hemoglobin, often present in patient samples. Since serum levels of bilirubin and hemoglobin are not predictable from patient to patient, an assay that is prone to interference with these molecules will not provide a robust clinical test that is reliable for all patients.
It is therefore desirable to provide a rapid acetaminophen assay that is accurate and reliable in the presence or absence of NAC, and which is less expensive than conventional immuno-based assays. It is also desirable to provide such an assay which is also less susceptible to interference with biological molecules present in patient samples compared to known assays.